Overcoming the Dark Side : Consequences of Destructive Leadership, and the Moderating Role of Hardiness

(1)

Overcoming the Dark Side

Consequences of Destructive Leadership, and the Moderating Role of Hardiness

Erik Spradbrow

Master Thesis

Norwegian Defence University College

2016

(2)

(3)

Acknowledgements

This thesis was written as part of the Staff Course at the Norwegian Defence University College in the autumn of 2016. The three sub-factors of hardiness will be described in the following pages, but they can also be used to categorise my acknowledgments.

Commitment. I would like to thank my supervisor, Tommy Krabberød for sound and helpful advice, and for the commitment it must have taken to read my first drafts.

I would also like to thank my wonderful wife for being committed to all aspects of our life and family, and for listening patiently when I was eager to explain the results of the work on this study. To the little ones; thank you for displaying true commitment and ensuring that no day was entirely filled by work on this thesis.

I also wish to thank Jan Tore Johansen, Geir Kolstø, Nick Sitter, Barry Spradbrow and Peter Spradbrow for providing comments and suggestions that have improved this study.

Challenge. Bartone (1999) states that hardiness is developed early on in life (Bartone, 1999).

I want to thank my parents for making sure that I was given challenges, and allowed to find my own way of solving them from an early age. Your efforts through the last decades ensured that I still enjoy challenges.

To the few destructive leaders I have worked for; thank you for the challenges that have made me realise that this topic was worth exploring. To the many exceptional leaders and

colleagues I have worked with; thank you for making this job so challenging - in a good way.

Control. Several people who have supported me have shown complete control and

understanding of these subjects. At the Norwegian Defence University College, I owe special thanks to Rino Bandlitz Johansen who provided the data, and always welcomed discussions.

Thank you for gently pushing me into the quantitative world. It has been great fun! Research for this study has also been greatly simplified by the outstanding service shown by the helpful librarians: Tamara, Per and Nina.

To Robert Buch, Jørn Hetland, Ole Christian Lang-Ree and Monica Martinussen; thank you for sharing your insights on methods, SPSS and JD-R in the early stages of this work.

Ancient roman cartographers used to write: here are lions (HIC SVNT LEONES) when denoting unknown territories on maps. Without this assistance this journey would have had considerably more lions.

(4)

Abstract

This study has examined if psychological hardiness, an established personal characteristic, can mitigate the consequences of destructive leadership behaviour amongst mid-level leaders in the Norwegian Armed Forces. To explore this, I have examined how destructive leadership behaviour was related to exhaustion and work engagement, and the moderating role of

hardiness on these relationships. The consequences for mission command, the current leadership philosophy in the Norwegian Armed Forces, have been discussed.

Data for this study was collected from officers and civilians who attended staff courses and leadership development programmes at the Norwegian Defence University College. The study used the job demands-resources model as a basis for the research - where destructive

leadership behaviour was used as a job demand; hardiness as a job resource; and exhaustion, and work engagement as dependent variables.

Findings indicated that superiors who displayed destructive leadership behaviour caused increased exhaustion, and reduced work engagement amongst the respondents. However, the analyses showed that the individuals who scored high on hardiness were less affected by the destructive behaviour that those who scored low on hardiness. This is in line with the theory of hardiness, and indicated that hardiness moderates the relationship.

The relationship between hardiness and destructive leadership behaviour has, to my

knowledge, not been studied before, and this study can therefore not only be of interest to the Norwegian Armed Forces, but also for others that are researching destructive leadership behaviour. This study also added insight to recent studies on destructive leadership in the Norwegian Armed Forces.

This study indicated that hardiness can be considered as a means, either through training or selection, to counter the negative outcomes of destructive leaders, and can therefore

contribute to a reduction in the negative consequences of destructive leadership.

Keywords: Destructive leadership behaviour, abusive supervision, toxic leadership, hardiness, engagement, exhaustion, JD-R, mission command, Norwegian Armed Forces, military

(5)

Sammendrag

I denne studien har jeg undersøkt om psykologisk hardførhet, en etablert

personlighetskarakteristikk, kan moderere forholdet mellom destruktiv lederadferd og utmattelse, og mellom destruktiv lederadferd og jobbengasjement hos mellomledere i Forsvaret. For å undersøke dette har jeg først sett på hvordan destruktiv ledelse påvirker de ansattes grad av utmattelse og jobbengasjement. Deretter har jeg undersøkt om hardførhet påvirker disse forholdene. Konsekvensene for Forsvaret og for Forsvarets ledelsesfilosofi, oppdragsbasert ledelse, har blitt diskutert.

Datagrunnlaget for undersøkelsen er samlet inn fra offiserer og sivile som har gått stabsstudiet eller har deltatt på et lederutviklingsprogram, LUPRO, ved Forsvarets Høgskole. Studien tar utgangspunkt i jobbkrav-ressursmodellen (job demands-resources model JD-R) som basis for forskningsmodellen der destruktiv ledelse brukes som et jobbkrav, hardførhet som en

jobbressurs og utmattelse og jobbengasjement som utfallsvariabler.

Funnene i studien indikerer at det å ha overordnede som utviser destruktiv ledelse fører til at mellomlederne både blir utmattet, og får redusert jobbengasjement. Studien viser imidlertid at for personer med høy grad av hardførhet, er disse følgene mindre enn for personer med lav hardførhet. Dette indikerer at hardførhet modererer disse relasjonene.

Forholdet mellom hardførhet og destruktiv ledelse har, så vidt jeg vet, ikke blitt studert før og denne studien kan derfor være av interesse for både Forsvaret og andre som studerer

destruktiv ledelse. Studien bidrar dermed også til forskning på destruktiv ledelse i Forsvaret og indikerer at hardførhet kan sees på som et verktøy, enten gjennom trening eller seleksjon, for å redusere konsekvensene av destruktive ledere, og er dermed et bidrag til hvordan man kan håndtere destruktive ledere – et fagfelt det er lite forskning på.

Nøkkelord: destruktiv ledelse, destruktiv lederadferd, hardførhet, robusthet, jobbengasjement, engasjement, utmattelse, utbrenthet, jobbkrav-ressursmodellen, JD-R, oppdragsbasert ledelse, Forsvaret, militæret

(6)

1 Introduction

“A diamond is just a piece of coal that handled stress exceptionally well” (Unknown).¹

With the emerging research on the destructive, and darker side of leadership in the Norwegian Armed Forces (Andersen, 2016; Forsvaret, 2014; Platek, 2015), the outcomes and prevalence of destructive leadership in an organisation that has long traditions of leadership development is concerning. This is even more so as the catchy title “Bad is stronger than good”

(Baumeister, Bratslavsky, Finkenauer, & Vohs, 2001) has become a well-used description of how destructive leadership has a stronger impact on subordinates than constructive leadership.

But is bad always stronger than good in military leadership? Or are there personal

characteristics amongst the officers that can reduce the impact of destructive leaders? This study will look at the outcomes of destructive leadership behaviour, and if psychological hardiness, a characteristic that predicts how individuals behave in challenging work environments (Maddi, 2004), can moderate these relationships. To relate this to the introductory quote, this study will explore whether hardiness is a characteristic that can predict if the employee will be just another piece of coal, or a potential diamond, when facing the dark sides of leadership.

The prevalence of destructive leadership has emerged through extensive research in the last three decades, and these studies indicate a proportion of leaders in many work environments who behave destructively towards employees, or the organisations interest – or both. A significant prevalence of destructive leaders has also been found in the Norwegian Armed Forces (Forsvarsstaben, 2016). Destructive behaviour by a superior creates a dysfunctional work environment, and causes a wide range of negative consequences for subordinates, including emotional exhaustion, reduced work engagement (Leary et al., 2013), stress and frustration (Ashforth, 1997). Destructive leadership can also affect the goals and interests of an organisation (Einarsen, Aasland, & Skogstad, 2007). While studies continue to show the destructiveness of bad leaders, it appears to be difficult to change their behaviour (Ulmer Jr et al., 2004), and it is therefore possible that a better solution than changing these, lies in

1 A number of internet sites attribute a similar quote to Henry Kissinger, but an exact reference for either of these quotes has not been found

(8)

attempting to increase the resilience of subordinates to reduce the consequences of this behaviour.

How people react to stressors has been shown to be related to the personal characteristics of individuals (Skomorovsky & Sudom, 2011). One such characteristic that has been found to predict how individuals behave and react, both mentally and physically, is psychological hardiness (e.g., Hystad, Eid, Laberg, Johnsen, & Bartone, 2009; Kobasa, 1979; Maddi &

Harvey, 2006). The hardiness construct was found when individuals with similar personal characteristics, later defined as hardiness, experienced high degrees of stress without suffering from the same illnesses that the individuals with low hardiness experienced (Kobasa, 1979).

Maddi (2002) later found that these individuals saw stressful situations as challenges, and as opportunities for personal growth. These hardy individuals attempted to actively take part and control events in their surroundings, were motivated by challenges, and were committed to their work and to life in general (Maddi, 2002). A potential value of hardiness for the

Norwegian Armed Forces is that studies have shown that hardiness can be trained (Bartone &

Hystad, 2010; Maddi, 2007), and that individuals with high hardiness values increase the hardiness of other individuals in an organisation (Bartone, 2006).

I will use the job demands-resources model (JD-R) as a base for the research model in this study. The JD-R model is a job-stress model that describes how personal and organisational demands, and resources lead to individual strains and well-being that affect the work

environment (Bakker & Demerouti, 2007). In this study I will consider destructive leadership behaviour to be a job demand, and hardiness a personal job resource. The dependent variables that will be explored are exhaustion, and work engagement.

The purpose of this study is to examine if psychological hardiness can be a characteristic that mitigates the negative effects caused by destructive leaders on mid-level leaders in the Norwegian Armed Forces. To examine this, I will look at how destructive leadership behaviour affects subordinate exhaustion and work engagement levels, and how hardiness moderates, or influences, these relationships. This will be examined in light of the impact these factors can have for the Norwegian Armed Forces, and for the leadership philosophy;

mission command.

In order to assess this I will use a cross-sectional quantitative design, with data collected from officers and civilians attending staff courses, and leadership development programmes from

(9)

2013 to 2015 at the Norwegian Defence University College. Using established scales, the respondents will rate statements that measure their own hardiness, and their perception of their previous leader’s behaviour. The respondents will also rate statements measuring their own levels of exhaustion and work engagement.

Some key articles² and numerous meta-analyses are the starting point for this study, and these will be examined in the literature review. In addition to working out from the reference lists of these articles, searches were also conducted on EBSCO,³ Google Scholar,⁴ CRIStin⁵ and PubSych.⁶ As there are several definitions and names given to destructive leadership,⁷ searches were conducted on the most common names of the construct.⁸

1.1 Mission command in the Norwegian armed forces

While destructive leadership can be a challenge for any organisation, the decentralised

philosophy of the Norwegian Armed Forces makes the hierarchical organisation vulnerable to factors that can reduce the engagement, and well-being of its subordinate leaders. The Chief of Defence Norway’s⁹ principle document on leadership (2012) states that the Norwegian Armed Forces previously relied on an order based leadership philosophy, where loyalty to assigned orders was more important than individual initiative, and assessments of the situation. Mission command¹⁰ was introduced as the Norwegian Armed Forces leadership philosophy in 1995, and acknowledges that the leaders with the best situation awareness are also best qualified to make decisions for their units based on the intentions of the order and their superiors (Forsvarssjefen, 2012).

The decentralised philosophy is based on delegating responsibility to the leader with the best overview of the situation, regardless of his or her position in the hierarchy (Forsvarssjefen, 2012). Mission command therefore relies on initiative, and on the offensive orientation of all leaders (Ben-Shalom & Shamir, 2011), as well as the relationship between leaders, and the

2 These include Tepper (2000) on destructive leadership, Maddi (2002) on hardiness, Bakker and Demerouti (2007) on the JD-R model and Maslach, Jackson and Leiter (1997) on exhaustion and work engagement.

3 https://www.ebsco.com/

4 https://scholar.google.no/

5 CRIStin is a Norwegian data base of Norwegian publications and journals (http://www.cristin.no/)

6 A database with contributions from Germany, France, Spain, Norway, USA and The Netherlands (https://www.pubpsych.eu/)

7 While I will use the term destructive leadership, I use the terms used in other studies when citing these

8 These include destructive leadership, abusive supervision, toxic leadership and petty tyranny

9 The updated translation of the title of the commander of the Norwegian Armed Forces

10 The Norwegian term is oppdragsbasert ledelse that freely translates to mission based leadership

(10)

individual engagement of leaders at all levels (Forsvarssjefen, 2012). The performance of the Norwegian Armed Forces is therefore dependent on the engagement and well-being of individual leaders, factors that destructive leadership behaviour has been shown to reduce amongst subordinates (Zhang & Liao, 2015).

If superiors create a work environment that reduces the individuals work engagement, the consequences may be that initiative is reduced amongst these subordinate leaders. This may also lead to reduced performance, or engagement, down through the chain of command, as these other subordinates depend on the vigour and initiative of their leaders. As the

philosophy is based on decentralisation, its success depends on the engagement, and vigour of subordinate leaders (Shamir, 2011). This is important not only in conflicts and combat

operations, but can affect the Norwegian Armed Forces in daily operations, and during normal service since the philosophy is valid in peace-time, conflict and war (Forsvarssjefen, 2012).

In order for mission command to be successful, superiors must trust their subordinates to make the right decisions (Ben-Shalom & Shamir, 2011). Gagné and Deci (2005) found that the support given by superiors, including acknowledging the decisions of subordinates, and encouraging self-initiation, was related to employee satisfaction and positive work-related attitudes (Gagné & Deci, 2005). To successfully make and execute these delegated decisions officers need psychological robustness in order to be able to think clearly, and make good decisions in complex environments (Forsvarssjefen, 2012).

Thus; as destructive leadership behaviour can affect the trust, engagement, and well-being of subordinate leaders, this behaviour is a potential threat to the success of mission command in the Norwegian Armed Forces.

1.2 Research question

Mission command is a decentralised leadership philosophy that relies on the individual leaders making decisions based on their understanding of the situation. Having engaged and vigorous leaders at all levels is therefore essential (Forsvarssjefen, 2012). From the results of three decades of research (Tepper, 2000; Zhang & Liao, 2015), it appears that destructive leadership behaviour could potentially lead to exhaustion, and reduced work engagement of these mid-level leaders. Finding factors that can moderate this negative relationship is therefore important for the successful use of mission command.

(11)

Hardiness is a characteristic that has been shown to predict how individuals cope with stressors (Maddi, 2002), and as destructive leadership can be considered a stressor to

subordinates (Tepper, 2000), hardiness appears to be a characteristic that has the potential to predict the outcomes of this behaviour. In order to examine this, the following research question is stated:

Can hardiness affect the outcomes of exposure to destructive leadership?

1.3 The significance of this study

While there is significant research on both destructive leadership behaviour and hardiness, I have not been able to find any studies that examine relationships between hardiness and destructive leadership. As hardiness is a characteristic that has been found to be trainable (Bartone & Hystad, 2010; Maddi, 2007) this study can contribute to finding ways of

improving work environments where destructive leaders are present, another area of research that is not well-studied (Einarsen, Skogstad, Rørvik, Lande, & Nielsen, 2016). This is

supported by a meta-analysis which found that out of 5364 studies on workplace

mistreatment, only twelve looked at how to intervene with the dysfunctional relationship (Hodgins, MacCurtain, & Mannix-McNamara, 2014), and none of these included hardiness.

Results from this study could therefore also be of interest for communities outside the Norwegian Armed Forces.

As mission command is vulnerable to dysfunctional leadership relations, and the reduced efficacy of individual leaders (Forsvarssjefen, 2012), it is important to find ways to reduce the impact of destructive leaders. While hardiness is not a personality characteristic that is trained in the Norwegian military academies today (Hystad, Olsen, Espevik, & Säfvenbom, 2015), this study could identify a system of training to counter the effects of destructive leaders.

The results of this study could also provide an insight on the relationship between destructive leadership, job demands and engagement. There has so far been little research that covers how job demands relate to engagement (as stated by Crawford, LePine, & Rich, 2010), and also how destructive leadership affects engagement (as stated by Scheuer, Burton, Barber, Finkelstein, & Parker, 2016).

A study on the outcomes of destructive leadership behaviour has been conducted on some of the respondents in this study previously with a recommendation that further research on larger

(12)

populations were required, and suggested using acknowledged models to better understand destructive leadership behaviour (Platek, 2015). This present study contributes to the previous research by using the JD-R model as a base for this study’s research model, the population has more than doubled, and I will include hardiness as a job resource that could potentially mitigate the outcomes of destructive leadership behaviour.

1.4 Delimitations

This study is a cross-sectional quantitative study, and the objective is to explore the relationships between variables. The causality of the relationships, that a longitudinal or qualitative study could provide (Creswell, 2014), has not been investigated. While destructive leadership behaviour has been shown to cause a number of negative consequences (for an overview, see Zhang & Liao, 2015), this study will focus on exhaustion and work

engagement, as these are two of the main outcomes in the JD-R model, and are important characteristics for leaders in the Norwegian Armed Forces.

Although the direct relationships between hardiness, and exhaustion and work engagement are interesting to examine, this study will only focus on how hardiness can be used to

moderate the relationship between destructive leadership behaviour, and the outcomes of this behaviour. This is done as the intention is to find ways that the Norwegian Armed Forces can counter destructive leadership, and not only look at the characteristics of the variables.

I have chosen to use the revised JD-R model (Bakker & Demerouti, 2007) as a base for the research model as it has strong empirical support (Schaufeli & Taris, 2014), and includes job demands and job resources, as well as strains and motivation – the factors that are necessary to explore the relationships in this study. The JD-R model has evolved since it was presented in 2001, and recent studies have proposed further expansions (Crawford et al., 2010).

However; the version from 2007 addresses the relationships that are required for this study.

This study reviews different sub-factors of destructive leadership behaviour, but this

behaviour will be treated as an overall construct in the analyses. Analyses of the relationships and consequences of the different forms of destructive leadership are therefore not examined.

The respondents are both military and civilian mid-level leaders, but will be examined as one group since the principle document on leadership in the Norwegian Armed Forces states that mission command applies to all leaders in the organisation (Forsvarssjefen, 2012).

(13)

2 Literature review

This chapter will start with a description of the JD-R model; the basis for this study’s research model. I will then present destructive leadership behaviour and hardiness; factors that are included in the research model as a job demand and a job resource; respectively. Following this, I will review the dependant variables: exhaustion and work engagement. The hypotheses that operationalise the research question will also be presented during this review. I will then present the research model depicting the relationships and hypotheses.

2.1 The job demands-resource model

The Norwegian Armed Forces relies on good interpersonal relationship and engaged leaders (Forsvarssjefen, 2012). For the success of mission command, a healthy work environment is therefore required. An established model for describing work environments, and the

interactions between factors that affect this, is the JD-R model (figure 1). Bakker and Demerouti (2007) found that these factors, that are positively or negatively associated with job stress and motivation, can be divided into two general categories; job demands and job resources. These can in turn lead to either job strains, or motivation and engagement amongst employees (Bakker & Demerouti, 2007). The JD-R model has strong empirical support, and has been used to predict numerous job demands, job resources and their outcomes (Schaufeli

& Taris, 2014).

Figure 1 The Job Demands-Resource model (Bakker & Demerouti, 2007)

(14)

Job demands require psychological efforts from employees to handle, and are therefore linked with a psychological cost (Schaufeli & Taris, 2014). The demands can in turn directly lead to strains such as burnout and exhaustion (Bakker & Demerouti, 2007). One of these job

demands is negative interpersonal relations (Schaufeli & Taris, 2014), and this can be caused by destructive leadership behaviour (Zhang & Liao, 2015). Destructive leadership behaviour has been found to cause strains on the subordinate. These strains include reduced motivation, well-being and job-satisfaction amongst subordinates (Einarsen et al., 2007). As the outcomes of job demands are similar, and as interpersonal relationship have been included in the JD-R model, and have been shown to be related to destructive leadership, destructive leadership behaviour will be treated as a job demand in this study.

Job demands can be further divided into challenge, and hindrance demands, based on how the demands are perceived by those experiencing them. While challenge demands can lead to positive emotions that in turn lead to active and problem-focused coping, hindrance demands can lead to emotion-focused coping, and decreased engagement (Crawford et al., 2010).

While it could be perceived that a demand is either a challenge, or a hindrance job demand, how individuals respond to a job demand is affected by both organisational and personal resources. Job resources are factors that affect how an individual perceives the work environment, including the job demands (Bakker & Demerouti, 2007). Job resources are physical, psychological or social aspects of the job such as support from colleagues, supervisory coaching and social relations in the organisation (Schaufeli & Bakker, 2004).

These resources can stimulate personal development, growth and organisational commitment, and are directly linked to increased work engagement for employees. They can also reduce the effects of job demands, as well as the psychological costs required to counter these demands (Hakanen, Schaufeli, & Ahola, 2008). In addition to the organisational resources, job

resources also include the personal characteristics of the individuals. These predict a person’s ability to influence their own work environment, and are associated with resilience (Schaufeli

& Taris, 2014). The personal resources described in the model have strong similarities with how psychological hardiness is described (Kobasa, 1979), and Bonanno (2004) found that hardiness was one of the traits that lead to resilience. This study therefore examines hardiness as a job resource.

(15)

While this shows that the model could be used to present destructive leadership behaviour, hardiness, and the outcomes of these factors, the model is significant for this study as it also describes how different processes connect these factors and outcomes. Bakker and Demerouti (2007) propose that the JD-R model consists of two different psychological processes caused by job demands and job resources; the health impairment, and the motivational process. The health impairment process describes how job demands can lead to job strains, such as burnout and exhaustion, while the motivational process describes how job resources can lead to employee motivation and engagement (Bakker & Demerouti, 2007).

In addition to the direct effects these processes describe, indirect effects where the two interact are also a part of the model. These effects describe job resources as a buffer on the strains caused by job demands, and that job demands interact with job resources when predicting employee engagement and motivation. By including both the direct and indirect effects, they propose that outcomes of various levels of demands and resources can be predicted as shown in figure 2 (Bakker & Demerouti, 2007).

Figure 2 The effects of job demands and job resources on strain and motivation (Bakker & Demerouti, 2007, p. 317)

While the model is flexible as to what factors can be included as outcomes of job demands and job resources (Bakker, Demerouti, & Sanz-Vergel, 2014), the two main outcomes that are

(16)

commonly studied are burnout/exhaustion and engagement (e.g., Crawford et al., 2010;

Hakanen et al., 2008; Martinussen et al., 2011; Schaufeli & Bakker, 2004). As mission command relies on initiative from subordinate leaders, including the mid-level leaders in this study, having exhausted or unengaged leaders will have severe consequences for the efficacy or performance of an organisation using a decentralised leadership philosophy. In this study, work engagement will be examined as a motivational outcome. As exhaustion is considered the key aspect of burnout (Maslach et al., 1997), this study will include exhaustion, as well as engagement, as the outcome variables.

From the predictions in figure 2, the expected outcomes of the interaction between hardiness, destructive leadership behaviour, exhaustion and work engagement in this study can be described (figure 3). When low-hardy individuals are exposed to destructive leadership behaviour, they are expected to experience exhaustion and reduced work engagement (high strain, low motivation), while the individuals with high hardiness are expected to experience average exhaustion, and still be engaged in the same circumstances (average strain, high motivation).

Figure 3 The effects of destructive leadership behaviour and hardiness on exhaustion and engagement

(17)

2.2 Destructive leadership behaviour

While military leadership is, at times, conducted under extremely demanding circumstances – such as combat situations, leadership in the armed forces is mainly not conducted under incoming enemy fire, but in normal work environments. The definitions, traits and

characteristics of civilian leadership, including the destructive forms, are therefore applicable when studying military leadership. For the Norwegian Armed forces, and for research on the Norwegian Armed Forces, the focus is still mainly on constructive leadership (see

Forsvarssjefen, 2012). This also previously applied to the general research on leadership, although research on destructive leadership has emerged in the last three decades (Einarsen et al., 2007).

Some definitions of leadership only describe behaviour that leads to better or more efficient organisations (Skogstad & Einarsen, 2004), and that the negative behaviour of leaders is something different from leadership – implying that the term leadership is only behaviour that has a positive effect on subordinates (Schilling, 2009). These opinions are perhaps the reasons why researchers use a wide range of names for this type of behaviour, including destructive leadership (Einarsen et al., 2007), petty tyranny (Ashforth, 1994), toxic leadership (Lipman- Blumen, 2005) and abusive supervision (Tepper, 2000). The wide range of behaviour that can be characterised as destructive was shown by Schilling (2009) when he defined the most typical aspects of negative leadership using both qualitative and quantitative methods. This study proposed eight categories¹¹ of destructive leadership behaviour, each with several subcategories (Schilling, 2009).

While categories of destructive leadership help to understand the types of behaviour that negatively affect subordinates, how this behaviour is conducted is also important for

understanding this form of leadership. Tepper (2000) defines this behaviour to be a “sustained display of hostile and nonverbal behaviours, excluding physical contact” (Tepper, 2000, p.

178).

11 a) Insincere leadership b) Despotic leadership c) Exploitative leadership d) Restrictive leadership e) Failed leadership f) Avoiding Leadership (active) g) Avoiding Leadership (passive) h) Laissez-faire leadership

(18)

To understand the consequences both for the subordinates, and the Norwegian Armed Forces, Einarsen and colleagues’ (2007) definition is useful as it includes how these leaders affect both individuals and organisations:

The systematic and repeated behaviour by a leader, supervisor or manager that violates the legitimate interest of the organisation by undermining and/or sabotaging the organisation's goals, tasks, resources, and effectiveness and/or the motivation, well- being or job satisfaction of subordinates. (Einarsen et al., 2007, p. 208)

This definition encompasses both organisational and interpersonal behaviour, while also showing that a leader may behave differently against the interests of these. While the definition specifies that the behaviour is required to be systematic and repeated to be considered destructive, it does not indicate that a leader is always constructive, or always destructive. On the contrary, leaders seldom behave destructively or constructively in all situations, and that the same leader therefore causes desirable as well as undesirable effects (Padilla, Hogan, & Kaiser, 2007). As mission command is based on trust and cohesion (Forsvarssjefen, 2012), building trust can be challenging even if a leader only behaves destructively in some situations. The main factors that define the leadership culture for Norwegian mission command include cohesion, trust, responsibility, professional ability, and being a role-model (Forsvarssjefen, 2012) - factors that a destructive leader is likely to affect through his or her behaviour.

It can also be presumed that a military leader sometimes considers behaviour that can be perceived as destructive to be necessary to carry out an order or conduct a mission. Steele (2011) found that leaders who regularly behaved destructively did so regardless of whether the behaviour was perceived to be necessary based on the given situation or not.

A conceptual model that encompasses both the subordinate and organisational perspective, was proposed by Einarsen and colleagues (2007) (figure 4), and defines destructive leadership behaviour in three categories; derailed, tyrannical and supportive-disloyal leadership - in addition to constructive leadership (Einarsen et al., 2007). The names used in the categories have also been used by other researchers both before, and after the model was presented.

(19)

Figure 4 A model of leadership behaviour (Einarsen et al., 2007)

Supportive-disloyal leadership includes leaders who behave in accordance with the interests of the subordinates, but have separate or diverging goals from the overall organisation (Einarsen et al., 2007).

Derailed leaders show no concern for either subordinates, or the organisation (Einarsen et al., 2007), and will often use charisma or their personal status to promote own interests, or personal gain (Aasland, Skogstad, Notelaers, Nielsen, & Einarsen, 2010).

Tyrannical leadership describes leaders who place the interest of the organisation above the well-being of their subordinates. While these leaders have the potential to produce exceptional results, and reach the goals of the organisation, it is done at the cost of the subordinates and employees (Ma, Karri, & Chittipeddi, 2004).

The model (figure 4) describes how active destructive behaviour affects the organisation or individuals. Since mission command is based on initiative, and relies on the offensive

orientation of all leaders (Ben-Shalom & Shamir, 2011), mission command is also vulnerable to passive leaders. A passive leader is a leader who does not fulfil his or her responsibilities (Aasland et al., 2010), and who offers little or no support, or guidance, to his or her

subordinates (Einarsen et al., 2007). Passive leadership, such as laissez-faire leadership¹², can

12 Laissez-faire leadership is a form of leadership where the leader does not meet the subordinates’ legitimate expectations, and involves an absence of leadership (Skogstad, Einarsen, Torsheim, Aasland, & Hetland, 2007)

(20)

be considered as destructive leadership behaviour since the lack of engagement can affect the interests of the organisation, or other employees (Einarsen et al., 2007).

Since mission command is based on delegated responsibility, and acting on the intentions of superiors (Forsvarssjefen, 2012), a passive leader in the Norwegian Armed Forces can cause greater consequences than not just fulfilling his or her own responsibilities. The passive leader’s superiors rely on the decisiveness of all their subordinate leaders (Ben-Shalom &

Shamir, 2011), and the subordinates of the passive leader also require intentions to act on.

This leadership form can therefore have severe implications for the Norwegian Armed Forces, as well as for the initiative and offensive orientation of other leaders in the organisation. This is supported by a study that found that passive leadership, including laissez-faire leadership, has the potential of leading to reduced performance, quality of work and work engagement (Skogstad et al., 2007). The consequences of laissez-faire leadership for the Norwegian Armed Forces have also been shown in a recent study, where this leadership style was found to lead to reduced organisational commitment, motivation and job satisfaction, as well as increased stress, and role conflicts amongst followers (Andersen, 2016). The study also supported that passive leadership could be perceived as destructive leadership behaviour.

Several studies include passive, and laissez-fair leadership, as destructive leadership

behaviour (e.g., Aasland et al., 2010; Einarsen et al., 2007; Schilling, 2009), but this view is not supported amongst all researchers. An opposing view is that passive forms of leadership behaviour should not be considered as destructive leadership, as the active forms of bad leadership are more severe, and have qualitatively different consequences for the followers (Schyns & Schilling, 2013). There is however agreement that destructive leadership

behaviour, and passive leadership has negative consequences for the work environment for the subordinates. Based on the severity of passive leaders for mission command, and that both passive leadership and destructive leadership behaviour causes negative consequences for subordinate, passive leadership will be included as part of destructive leadership behaviour in this study.

The above models and definitions apply to both civilian and military leaders. Due to the extreme environments where military leadership must be applicable, and the dependency on individual leaders in mission command, military leadership may differ from civilian

leadership in some aspects. A context-specific model for destructive leadership behaviour,

(21)

aimed at military personnel, was therefore developed (Larsson, Fors Brandebo, & Nilsson, 2012). By conducting a qualitative and quantitative study on Swedish military groups, Larsson and colleagues (2012) designed a model with five sub-categories of destructive leadership behaviour,¹³ including passive leadership, and a short scale to measure these factors (Larsson et al., 2012).

As introduced in the first chapter, the prevalence of destructive leadership has shown that this is a wide-spread problem. A meta-analysis has shown that between 11% and 18% of

employees (N = 130 973) reported to have experienced leaders who show signs of workspace bullying (Nielsen, Matthiesen, & Einarsen, 2010), and an estimated 14% of all employees in the US are affected by destructive leaders (Tepper, 2007). In a Norwegian study 60%

(N = 2539) reported to have been exposed to destructive leadership behaviour in the past six months (Aasland et al., 2010).

That the prevalence of destructive leaders amongst officers is large enough to be a concern for the Norwegian Armed Forces seems apparent through the latest Norwegian Defence Force employee survey. The survey showed that 32% of all respondents (N = 9249) had

experienced negative leadership behaviour (Forsvarsstaben, 2016). The Norwegian results are also comparable to other armed forces, and a study on American military service members (N = 8780 to 9580) found that 20% viewed their leaders negatively (Steele, 2011). The discrepancy in reported numbers between studies can partly be explained by how studies are designed, but also that leadership styles vary between countries (Nielsen et al., 2010).

However; even the lowest estimates show that a significant portion of leaders display destructive behaviour, and the 32% found in the Norwegian Armed Forces survey (Forsvarsstaben, 2016) is a significant amount.

In addition to the negative effects of destructive leadership behaviour mentioned, research has shown that destructive leadership leads to increased exhaustion and burnout (Tepper, 2000) and work-family conflicts (Carlson, Ferguson, Hunter, & Whitten, 2012), as well as reduced health (Nielsen & Einarsen, 2012), work engagement (Crawford et al., 2010; Einarsen et al., 2016), team performance (Calheiros, 2012), job satisfaction (Nielsen & Einarsen, 2012;

Steele, 2011), psychological capital (Calheiros, 2012) and morale leading to job turnover

13 The categories are described as: a) arrogant and unfair b) threats, punishments and over-demands c) ego- oriented and false d) passive and cowardly e) uncertain, unclear messy (Larsson et al., 2012)

(22)

(Steele, 2011). Destructive leadership behaviour has also been found to lead to subordinates having difficulties in understanding their role and tasks in the organisation, as well as negatively affecting how subordinate leaders perceive their own leadership and their organisations team-spirit (Maddi, 2002; Schilling, 2009).

Mission command relies on active subordinate leaders who are capable of making

independent decisions, and individuals with reduced levels of performance or engagement can negatively influence how mission command is executed. The review in this chapter indicates that destructive leadership behaviour can have a negative impact on this type of leadership philosophy. However; from the theory of the JD-R model, it can be presumed that job

resources can influence how damaging a job demand such as destructive leadership is for the mid-level leaders in this study. Hardiness is a personal job resource that has the potential to reduce the severity of outcomes of exposure to destructive leaders.

2.3 Hardiness

In looking into the relationship between stress and health, research on personality as a factor for keeping healthy led to the development of a theory on hardiness (Kobasa, 1979; Maddi, 2002). Psychologist Suzanna Kobasa (1979) found that these personality structures could predict individual outcomes of stressful environments.

The characteristics of hardiness were later investigated in an extensive twelve year

longitudinal study, which started in 1975, on 450 managers working at Illinois Bell Telephone (IBT). Psychologist Salvatore Maddi and colleagues conducted annual psychological and medical tests on the managers, but after the collapse of the company in 1981,¹⁴ the tests also examined the effects of stress on the individuals. Two thirds of the managers in the survey showed negative stress reactions as a consequence of the company’s collapse, and the reactions included suicide, divorce and physical and mental health issues. The study also showed that one third of the respondents not only coped with the collapse, and got through the hardship, but also thrived in the challenging circumstances. These managers had higher scores on hardiness, and felt enlivened, were promoted in either IBT or their new companies, and suffered fewer health issues than the managers with low hardiness scores (Maddi, 2004).

14 IBT reduced the number of employees from 26 000 to 14 000 in under one year, as a consequence of deregulation of the American telecom sector (Maddi, 2002).

(23)

These characteristics of hardiness have since been reproduced in numerous studies (for an overview, see Eschleman, Bowling, & Alarcon, 2010).

Hardy individuals saw that by taking control of the events, and not blaming others, or other circumstances, for their situation, they are capable of making decisions that lead to positive outcomes in stressful situations (Maddi, 2004). In addition, leaders with high hardiness levels have been found to not only resolve their own situations, but also increase the hardiness levels, and motivation of their organisations and subordinates - who in turn respond with more effective coping strategies (Bartone, 2006).

A hardy individual will see stressful and painful experiences as part of an interesting and worthwhile life, and can be described as open to challenges and changes (Bartone, 1999).

These individuals take control of events in their lives in order to influence their own situation, and are committed to their work, their surroundings, people and life in general (Maddi, 2002).

Leaders who scored high on stress, but low on illness showed higher values of control, commitment and challenge than leaders that scored high on both stress and illness (Kobasa, 1979). The conceptualisation of hardiness is therefore based on the these three facets:

commitment, control and challenge (Maddi, 2002).

Military organisations rely on cohesion and trust, and commitment to the organisation and their surroundings is also a facet of hardiness that describes peoples’ predisposition to taking an active part in the activities and decisions that are related to their own and their

organisations situation. Individuals who are strong in commitment are engaged in interpersonal relationships, and social activities (Kobasa, 1979), and are involved in the events around them; as opposed to withdrawing from the situation and becoming isolated and detached from decisions and activities (Maddi, 2002). Committed mid-level leaders can therefore be presumed to maintain cohesion to their organisation, even when their immediate superiors behave destructively. This is an important trait for mission command that relies on committed leaders who are expected to make decisions based on the intentions of superiors, and for the greater benefit of the organisation.

Control is related to a desire to influence events involving oneself, as opposed to becoming passive, and experiencing a feeling of powerlessness (Maddi, 2002). Kobasa (1979) described control as attempting to take active measures, and influence the outcome of decisions. These individuals also have the ability to incorporate stressful events into their own long term plan,

(24)

and choose their own courses of action on how to handle stressors. The ability of an

individual to cope with stressful situations is also included in control, as opposed to those who become ill, or incapacitated, by similar events (Kobasa, 1979). When faced with destructive leaders, influencing one’s own work environment by actively trying to resolve the

dysfunctional relationship can be a way that hardy individuals reduce the strains caused by destructive leaders.

Taking the initiative and accepting challenges, even when the option to choose an easier solution is present, is an important part of mission command, and this trait is also part of the challenge facet of hardiness. Individuals with high challenge scores see that the way to live an interesting life is through stressful, and potentially painful experiences, and not by choosing the easy, monotonous or routine options (Bartone, 1999; Maddi, Khoshaba, Harvey, Fazel, &

Resurreccion, 2010). Scoring high in challenge is related to believing that change is normal, and that these events are an opportunity to learn and grow from the experience (Maddi, 2007).

Those with low challenge scores will aim at avoiding change, threats and uncertainties (Maddi, 2002).

Hardiness has been found to be a relatively stable disposition, and a pattern of attitudes that enables people to make the most of stressful circumstances (Maddi, 2007). The disposition also provides courage to change stressors and potential personal disasters into opportunities for growth (Maddi et al., 2010). While hardiness is a stable disposition, it has been found to be a trainable attitude (Maddi, 2007), including amongst military personnel (Bartone &

Hystad, 2010).

A number of studies have examined the effects of hardiness on military personnel (e.g., Bartone, 2006; Carston & Gardner, 2009). Hardiness has been shown to be a characteristic that predicts adaptivity amongst military leaders (Bartone, Kelly, & Matthews, 2013), and a trait that can predict transformational leaders (Johnsen, Eid, Pallesen, Bartone, & Nissestad, 2009). Hardiness has also been shown to be related with the perception of conducting meaningful work (Britt, Adler, & Bartone, 2001), and to buffer negative aspects of the military profession, such as preventing burnout (Lo Bue, Taverniers, Mylle, & Euwema, 2013), predicting psychological well-being (Skomorovsky & Sudom, 2011), and reducing war-related stress and other psychological combat related injuries (Bartone, 1999; Maddi, 2007).

(25)

While Kobasa and Maddi are normally referred to as the conceptual developers of hardiness, the personality’s sub-factors are based on numerous theories by existential philosophers and psychologists dating back to 1947 (Kobasa, 1979). It is therefore not surprising that hardiness has been criticised for having properties that coincide with other personality constructs - such as the five factor model (Maddi & Harvey, 2006). Maddi (2007) also suggests that another explanation for this criticism is that the original hardiness scale only used negatively worded statements based on opposite existing scales, and the statements were criticised for being little more than negatively worded versions of statements that measure neuroticism, one of the factors in the five factor model (Maddi, 2007). The first studies were also criticised, as the challenge sub-factor was unrelated to the other two sub-factors in some studies (Maddi &

Harvey, 2006). As a new version of the scale, that specifically measured hardiness, was released shortly after the first versions, Maddi and Harvey claim that the issues were resolved (Maddi & Harvey, 2006). Sinclair and Tetrik (2000) later showed that hardiness is separable from neuroticism, and that hardiness is also a useful predictor of both health, and performance issues.

Another criticism of hardiness studies has been that a variety of scales have been used for measuring the construct; thus making comparisons difficult and unreliable (Klag & Bradley, 2004). This also appears to have improved as most studies today use versions of either the Dispositional Resilience Scale, as this study does, or the Personal Views Survey (Hystad, Eid, Johnsen, Laberg, & Thomas Bartone, 2010).

Based on this review of destructive leadership, and the characteristics of hardiness, it is expected that these factors can be considered to be a job demand and a job resource,

respectively. While these have been shown to predict, and affect, a wide range of outcomes, this study will focus on exhaustion and work engagement - as stated in chapter 2.1.

2.4 Exhaustion

Exhaustion¹⁵ is a psychological state where individuals feel a lack of energy, and that their emotional resources are depleted. This can in turn lead to frustration, as the individuals feel that they are not able to provide their previous commitment (Cordes & Dougherty, 1993).

Exhaustion is often considered to be the core dimension of burnout, and while burnout

15 Both the term exhaustion and emotional exhaustion is used in studies. In this study I will use exhaustion.

(26)

consists of the three sub-dimensions: exhaustion, depersonalisation and reduced personal accomplishment (Maslach et al., 1997), exhaustion is referred to as the first stage of burnout (Cordes & Dougherty, 1993). People describing themselves as burnt-out often refer to the state of being exhausted (Maslach, Schaufeli, & Leiter, 2001). Research on burnout and exhaustion started with studies on workers in the human services, but later studies have shown that this psychological state also appears in other occupational fields (Demerouti, Bakker, Nachreiner, & Schaufeli, 2001), including law enforcement (Martinussen, Richardsen, & Burke, 2007) and armed forces (Leiter, Clark, & Durup, 1994).

Maslach et al. (2001) found that exhaustion can lead to attempts to distance oneself

cognitively and emotionally from one’s work in order to handle the feeling of work overload (Maslach et al., 2001). This can in turn lead to reduced organisational commitment and turnover intentions (Cropanzano, Rupp, & Byrne, 2003). While engaged individuals are more active and positive to new experiences, exhaustion can lead to health problems, absenteeism and reduced performance (Bakker et al., 2014), and is a cause of inefficacy as it is difficult to be engaged in one’s work while feeling emotionally depleted (Maslach et al., 2001).

2.4.1 Destructive leadership behaviour and exhaustion

Mission command in the Norwegian Armed Forces is a leader-centric philosophy that relies heavily on the individual leader’s ability to take the initiative, and have good situation awareness. In the description of this philosophy; dedication and involvement are central personal characteristics for the successful implementation of mission command

(Forsvarssjefen, 2012). It therefore seems plausible that leaders who suffer from reduced efficacy, involvement, or who distance themselves from their tasks, as a consequence of exhaustion caused by destructive leadership behaviour, will potentially also reduce the

efficacy of their own- and supported units. As mission command relies on subordinate leaders to have sufficient psychological robustness to be able to make sound decisions in demanding environments (Forsvarssjefen, 2012), exhaustion amongst these could in turn lead to a reduction of the effectiveness of this decentralised leadership philosophy.

A number of studies have supported the relationship between job demands and burnout and/or exhaustion as stated in the JD-R model (e.g., Demerouti et al., 2001; Hakanen et al., 2008;

Schaufeli & Taris, 2014; Scheuer et al., 2016), and Bakker and Demerouti (2007) showed that high demands, and low job resources significantly predict burnout amongst employees.

(27)

That abusive supervision can be considered a job demand that causes exhaustion has also been shown in a recent study (Scheuer et al., 2016), and several studies have shown that destructive leadership is related to exhaustion (e.g., Carlson et al., 2012; Tepper, 2000, 2007).

Recent studies on the employees of the Norwegian Armed Forces showed a relationship between destructive leadership behaviour and exhaustion (Platek, 2015),¹⁶ and that laissez- faire leadership was related to stress (Andersen, 2016), that in turn can lead to exhaustion (Maslach et al., 2001).

From these findings, and based on the JD-R mode, I expect that exposure to destructive leadership behaviour can lead to exhaustion for the mid-level leaders in this study. In order to examine this relationship amongst the leaders in this study, the following hypothesis is stated;

Hypothesis 1: Destructive leadership behaviour leads to increased exhaustion amongst subordinates.

2.4.2 Hardiness as a moderator on the relationship between destructive leadership and exhaustion

Mission command relies on the subordinate leader’s initiative, and ability to think clearly in stressful environments (Forsvarssjefen, 2012) and exhausted leaders are not likely to take this initiative. As studies have shown that destructive leadership behaviour can cause exhaustion (e.g., Tepper, 2000), this ability can be reduced by exposure to destructive leaders. This can in turn lead to reduced efficacy of their units, and of mission command as a philosophy, as it relies on the initiative, and offensive orientation of all leaders (Ben-Shalom & Shamir, 2011).

While destructive leadership behaviour can be considered a significant job demand, the consequences of this behaviour varies amongst those exposed to it. Individual characteristics, and coping mechanisms, can be considered as moderating factors (Nielsen & Einarsen, 2012).

Thus, having a trait that can moderate this relationship can be positive for both the

subordinate leader’s vigour, as well as for superiors and subordinates of this leader (Bartone, 2006), as the Norwegian Armed Forces is dependent on the vigour of all its leaders

(Forsvarssjefen, 2012).

16 Although the study focuses on burnout, only the exhaustion dimension of MBI-GS was used and the results are therefore comparable.

(28)

Several studies show that hardiness is one of the characteristics that can moderate, or reduce, the chances of burnout, and thereby exhaustion (e.g.,DePew, Gordon, Yoder, & Goodwin, 1999; Henderson, 2015; Lo Bue et al., 2013; Maddi, 2004; Maslach et al., 2001). Similar results were also found in military combat operations, where hardiness predicted reduced exhaustion (Lo Bue et al., 2013). In a large study on Norwegian Defence Force employees (N = 15 410), hardiness was found to predict absence due to sickness, and that individual differences in hardiness predicted the amount of absence spells when job demands were high (Hystad, Eid, & Brevik, 2011).

The moderating effect of hardiness on the relationship between stress and burnout has also been shown, and hardiness explained 35% of the variation of burnout in one study (DePew et al., 1999). In a meta-analysis of 114 studies, Alarcon et al. (2009) found that the mean

negative relationship between hardiness and exhaustion (r = -.36) was stronger than the negative relationship between exhaustion, and several other personality traits, including the five factor model, proactive behaviour, optimism and type A behaviour (Alarcon et al., 2009).

Abusive supervision has also been found to only be related to increased exhaustion, and not a job demand that could also be a challenge for some individuals (Scheuer et al., 2016). This could potentially be a finding that contradicts the theory of hardiness which describes that individual hardiness levels predicts how a stressor is perceived (Maddi, 2002).

As hardiness has been shown to moderate stressors in numerous studies, and destructive leadership behaviour can be considered one of these stressors, I expect that while the mid- level leaders with low hardiness can become exhausted due to destructive leadership behaviour, the hardy individuals can perceive this same behaviour as a challenge, or find ways to control the situation, and thus suffer from less negative consequences. This is in line with Bartone and Hystad’s (2010) preposition that the hardy individuals still perceive

stressors, but do not show the same symptoms, or the same decrease in performance, as the low-hardy individuals (Bartone & Hystad, 2010). To examine these relationships on the respondents in this study, the following hypothesis is stated:

Hypothesis 2: Hardiness will moderate the relationship between destructive leadership behaviour and exhaustion

(29)

2.5 Work Engagement

While employee engagement, and work engagement are well used terms in business, and amongst consultants, few research articles explore the topic, and those that do were produced in the last 16 years (Schaufeli & Bakker, 2010).¹⁷ While exhaustion can be characterised as a lack of energy that can lead to reduced commitment and efficacy, work engagement can be defined by vigour, dedication and absorption (Schaufeli & Bakker, 2004), and has been found to include physical, cognitive and emotional engagement to work (Babcock-Roberson &

Strickland, 2010).

The level of work engagement amongst employees varies based on individual perceptions of the benefits and meaningfulness of their work, and on the individual resources they perceive that they can contribute with (Kahn, 1990). Thus, work engagement is described as a positive, and work-related state of mind (González-Romá, Schaufeli, Bakker, & Lloret, 2006).

Engaged individuals are characterised as having high energy levels, a willingness to be involved in their work, while being concentrated, and feeling that one’s work is significant (Schaufeli & Bakker, 2010). This leads to these individuals taking more initiative, and creating their own positive feed-back, as well as being energetic and enthusiastic employees who identify with their work (Demerouti, Cropanzano, Bakker, & Leiter, 2010).

Some researchers view burnout and work engagement as pair-wise opposites, antitheses or consider work engagement to be the absence of exhaustion (e.g., González-Romá et al., 2006;

Lo Bue et al., 2013; Maslach et al., 1997). However; work engagement is seen by others as an independent construct that is negatively related to exhaustion and burnout (Schaufeli &

Bakker, 2004). Work engagement is considered a persistent state of emotions, as opposed to burnout which is described as a more momentary emotion (Schaufeli & Bakker, 2010). A recent Norwegian study showed a relationship between destructive leadership behaviour and exhaustion, but not with work engagement (Platek, 2015) - implying that the two factors are separate constructs. This is also supported by other studies that indicate that the concepts can be discriminated empirically (e.g., González-Romá et al., 2006; Schaufeli & Bakker, 2004;

Schaufeli, Taris, & Van Rhenen, 2008). This study therefore treats the variables as

17 A Google search on employee engagement and work engagement gave 645 130 hits on Google, bute only 1898 records in Google scholar (Schaufeli & Bakker, 2010).

(30)

independent states (González-Romá et al., 2006; Schaufeli & Bakker, 2004) - potentially influenced differently by destructive leadership behaviour.

2.5.1 Destructive leadership behaviour and work engagement

For any organisation an employee’s individual work engagement can affect how an

organisation functions, or performs. For the Norwegian Armed Forces, that is dependent on the initiative of leaders at all levels (Forsvarssjefen, 2012), the engagement of the mid-level leaders in this study are essential. Mission command is based on the engagement, and vigour of its leaders (Shamir, 2011), and if destructive leadership behaviour is found to reduce the work engagement of subordinates, this can lead to these mid-level leaders becoming less involved, and can in turn lead to subordinates with less initiative, and who are less offensively oriented - attributes that mission command relies on (Ben-Shalom & Shamir, 2011).

The JD-R model not only describes how job demands lead to employee strains, and job resources lead to employee engagement, but also how job demands contribute to reducing the work engagement for employees (Bakker & Demerouti, 2007).

The relationship between job demands and work engagement has, not always been found to be statistically significant (Schaufeli & Taris, 2014), and engagement has in some studies been shown to be exclusively predicted by job resources, and not job demands (Bakker, Demerouti, De Boer, & Schaufeli, 2003; Bakker, Demerouti, & Verbeke, 2004; Schaufeli &

Bakker, 2004).

Some job demands have been found to predict reduced work engagement (Crawford et al., 2010; Martinussen et al., 2011), but the described variation was less for engagement than the variation explained for exhaustion (Martinussen et al., 2011). This can indicate that

destructive leadership behaviour will explain more of the variation of exhaustion than of work engagement; however both the relationships were still found in Martinussen et al’s study.

In all, few studies have shown the relationship between job demands and engagement (Crawford et al., 2010), and the same also applies to destructive leadership and engagement (as stated by Scheuer et al., 2016).

While there are studies that have shown that destructive leadership does not reduce work engagement (Schaufeli & Bakker, 2004), including a recent study on Norwegian officers

(31)

(Platek, 2015), there is empirical support behind the proposal that destructive leadership behaviour does cause reduced work engagement (for an overview, see Zhang & Liao, 2015).

Both destructive leadership behaviour and passive leadership has been found to affect work engagement (Leary et al., 2013), and abusive supervision viewed as a hindrance job demand also showed this relationship (Poon, 2011).

While destructive leadership may have a stronger impact on the subordinate exhaustion levels than on the work engagement of these, there is both theoretical (Bakker & Demerouti, 2007), and empirical support (see Zhang & Liao, 2015) behind that destructive leadership behaviour causes reduced work engagement. As the consequences of this reduced engagement amongst the mid-level leaders can lead to a less effective implementation of mission command, this relationship will be examined by stating the following hypothesis:

Hypothesis 3: Destructive leadership behaviour leads to reduced work engagement amongst subordinates.

2.5.2 Hardiness as a moderator on the relationship between destructive leadership and work engagement

There are few studies that have shown a relationship between destructive leadership

behaviour and work engagement (Scheuer et al., 2016), and little research on how to reduce the negative outcomes of destructive leadership behaviour in general (Einarsen et al., 2016).

However; the JD-R model describes that job demands affect work engagement, and job resources are expected to moderate this relationship (Bakker & Demerouti, 2007). There are several reasons why I expect the relationship between destructive leadership behaviour and work engagement to be moderated by hardiness.

First, individuals who score high on hardiness are assumed to have the capability of using stressful situations to enhance their own psychological growth and performance (e.g., Maddi, 2007), and job demands appraised as challenges have been found to be related to work engagement - via the potential for personal growth (Einarsen et al., 2016). Amongst military personnel, hardiness has been found to be related to feeling engaged in their mission, finding their work meaningful (Britt et al., 2001), and to be a predictor of work engagement (Lo Bue et al., 2013). With hardiness showing these characteristics, I also expect individuals with high hardiness to be able to perceive a destructive leader as a challenge that does not reduce his or

Overcoming the Dark Side : Consequences of Destructive Leadership, and the Moderating Role of Hardiness